Body mass index is sufficient to predict diabetes or metabolic syndrome, Italian study finds

Body mass index (BMI) predicts the risk of developing diabetes or metabolic syndrome as reliably as more precise tests of body fat distribution in people with HIV, an Italian study presented last week at the 19th European AIDS Conference (EACS 2023) in Warsaw has found.

However, the study did not report on the ethnicity of participants or analyse changes in weight and body composition according to ethnicity, so it is not clear if the findings are applicable in all settings.

BMI (weight in kilograms divided by the square of the height in metres) can be used to predict the risk of developing diabetes or metabolic syndrome (any three of dyslipidaemia, raised glucose levels, high blood pressure or obesity).

But BMI doesn’t capture the distribution of fat or the extent to which muscle contributes to body mass. Visceral fat, which accumulates around the organs, plays a much greater role in the development of heart disease than subcutaneous fat under the skin. And in two people with the same BMI, the one with the higher proportion of fat-free mass (muscle and bone) will usually have a lower risk of diabetes and heart disease. This is probably because muscle tissue uses most of the glucose in the body, so more muscle means more glucose is taken out of circulation and burned for energy.

Ethnic differences in body fat distribution and lean muscle mass also contribute to differences in the BMI levels at which the risks of diabetes or cardiovascular increase. A recent study carried out in the United Kingdom found that people of South Asian descent with a BMI of 23kg/m2 had the same diabetes risk as White people with a BMI of 30kg/m2 (clinically obese).

However, body composition can only be measured by DEXA scan or MRI and these tests are not available in primary health care or many HIV clinics. Dr Jovana Milić and colleagues at the University of Modena HIV Metabolic Clinic in Italy wanted to find out whether changes in body composition measured by DEXA were better predictors than BMI of the development of diabetes or metabolic syndrome in people with HIV.

They looked at the predictive value of the two measurements in 1895 people with HIV followed for at least six months who did not have diabetes or metabolic syndrome at baseline. People with HIV who received care at the clinic were eligible for inclusion in the analysis if they had at least two clinic visits and sets of measurements between 2008 and 2022.

All participants in the prospective study underwent DEXA scans to measure body composition and bone mass.  

The study had two primary outcomes: development of diabetes or metabolic syndrome. Diabetes was defined as an HbA1c measurement of 6.5% or greater, or use of diabetes-specific medication, or fasting glycaemia >126mg/dL, or glycaemia >200mg/dL at any time. Metabolic syndrome was defined by at least three of the following measures: waist circumference of 102cm or above in men, 88cm or above in women; blood pressure >130/85mmgHg or blood pressure medication; fasting triglycerides of 150mg/dL or above or medication; fasting HDL cholesterol <40mg/dL or medication; fasting glucose of 100mg/dl or above, or medication.

The study also measured a composite outcome of diabetes or metabolic syndrome.

The study followed 1895 participants (70% male) for a median of 5.8 years, during which they underwent a median of five DEXA scans.

Participants were middle-aged (mean age 45 years) and the median body mass index was 23.2 kg/m2 (normal range). Participants had well-controlled viral load (88% undetectable) and high CD4 counts (median 625) at baseline.

During the follow-up, 219 people developed diabetes (incidence 1.5 per 100 person-years of follow-up), 377 developed metabolic syndrome (incidence of 3.7 per 100 person-years) and 417 developed the composite outcome (diabetes or metabolic syndrome (4.1 per 100 person-years).

In a model which controlled for insulin resistance, HIV-related characteristics, alcohol use, smoking and physical activity, changes in body mass index, trunk fat, lean body mass, lumbar bone mineral density and visceral fat, each predicted the development of the composite outcome. Increases in visceral fat were most strongly associated with development of the composite outcome.

But when the same analysis was carried out for diabetes alone, changes in trunk fat, lumbar bone mineral density and visceral fat were not significant predictors of diabetes.

Changes in all measures predicted metabolic syndrome, with changes in visceral fat proving the strongest predictor.

Find out more: Type 2 diabetes and HIV

The study investigators say that although visceral fat and lean body mass changes proved the best predictors of the composite outcome, changes in body mass index were also predictive. If DEXA scanning is not available, body mass index and changes in BMI are reliable for estimating the risk of diabetes and metabolic syndrome. The study investigators say that more research is needed to determine which measures most reliably predict the risk of cardiovascular events or non-alcoholic fatty liver disease in people with HIV.